{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/spectral-multigraph-networks-for-discovering","title":"Spectral Multigraph Networks for Discovering and Fusing Relationships in Molecules","arxiv_id":"1811.09595","date":"2018-11-23","proceeding":null,"authors":["Boris Knyazev","Xiao Lin","Mohamed R. Amer","Graham W. Taylor"],"abstract":"Spectral Graph Convolutional Networks (GCNs) are a generalization of\nconvolutional networks to learning on graph-structured data. Applications of\nspectral GCNs have been successful, but limited to a few problems where the\ngraph is fixed, such as shape correspondence and node classification. In this\nwork, we address this limitation by revisiting a particular family of spectral\ngraph networks, Chebyshev GCNs, showing its efficacy in solving graph\nclassification tasks with a variable graph structure and size. Chebyshev GCNs\nrestrict graphs to have at most one edge between any pair of nodes. To this\nend, we propose a novel multigraph network that learns from multi-relational\ngraphs. We model learned edges with abstract meaning and experiment with\ndifferent ways to fuse the representations extracted from annotated and learned\nedges, achieving competitive results on a variety of chemical classification\nbenchmarks.","url_abs":"http://arxiv.org/abs/1811.09595v1","url_pdf":"http://arxiv.org/pdf/1811.09595v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"spectral-multigraph-networks-for-discovering","repo_url":"https://github.com/bknyaz/graph_nn","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"graph-classification","task_name":"Graph Classification"},{"task_slug":"node-classification","task_name":"Node Classification"}],"methods":[{"method_slug":"graph-convolutional-networks","method_name":"Graph Convolutional Networks"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/graph-classification-on-enzymes","task":"Graph Classification","dataset":"ENZYMES","model":"Multigraph ChebNet","rank_in_archive_order":27,"of":54,"metrics":{"Accuracy":"61.7%"},"uses_additional_data":false},{"leaderboard":"/sota/graph-classification-on-mutag","task":"Graph Classification","dataset":"MUTAG","model":"Multigraph ChebNet","rank_in_archive_order":29,"of":74,"metrics":{"Accuracy":"89.1%"},"uses_additional_data":false},{"leaderboard":"/sota/graph-classification-on-nci1","task":"Graph Classification","dataset":"NCI1","model":"Multigraph ChebNet","rank_in_archive_order":27,"of":69,"metrics":{"Accuracy":"83.4%"},"uses_additional_data":false},{"leaderboard":"/sota/graph-classification-on-nci109","task":"Graph Classification","dataset":"NCI109","model":"Multigraph ChebNet","rank_in_archive_order":19,"of":38,"metrics":{"Accuracy":"82.0"},"uses_additional_data":false},{"leaderboard":"/sota/graph-classification-on-proteins","task":"Graph Classification","dataset":"PROTEINS","model":"Multigraph ChebNet","rank_in_archive_order":42,"of":103,"metrics":{"Accuracy":"76.5%"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1811.09595","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}